帮我把下面这段话按照行人重识别领域的论文的风格翻译成英文：E在等式1中表示图G的边集。EV表示由视觉信息构建出来的边集EST表示基于时空信息构建出来的边集。具体来说对于EV我们对来自训练集V的任意一张图像VI根据和其他所有图像的亲和性进行排序然后选取top k张图像构建EV。

In Equation 1, E represents the edge set of graph G. EV represents the edge set constructed from visual information, and EST represents the edge set constructed based on spatiotemporal information. Specifically, for EV, we sort any image VI from the training set V based on its affinity with all other images, and then select the top k images to construct EV. This is how we construct the edge set EV from visual information.